Go West: The Growth of Bollywood - Lorenzen, Mark - CBS Research Portal
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Go West: The Growth of Bollywood Lorenzen, Mark Document Version Final published version Publication date: 2009 License CC BY-NC-ND Citation for published version (APA): Lorenzen, M. (2009). Go West: The Growth of Bollywood. imagine.. CBS. Creative Encounters Working Paper No. 26 Link to publication in CBS Research Portal General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. Take down policy If you believe that this document breaches copyright please contact us (research.lib@cbs.dk) providing details, and we will remove access to the work immediately and investigate your claim. Download date: 12. Aug. 2021
Creativity at Work: Go West: The Growth of Bollywood By Mark Lorenzen April 2009 Page 1 / 38 Creative Encounters Working Paper # 26
Abstract This paper builds insight into how globalization impacts cultural clusters, through a case study of Bollywood, the Indian film cluster in Mumbai. The paper’s analysis of the recent growth and consolidation of Bollywood, as well as the cluster’s development of a new film formula, illustrates that globalization does not necessarily entail westernization of culture. Instead, the paper suggests that early-mover advantages held by the world’s core cultural clusters may be eroded by globalization, as it creates pipelines of information, talent and capital, allowing hitherto peripheral cultural clusters to access export markets and develop exportable products. Analyzing the role of the Indian diasporas for the export growth of Bollywood, the paper also offers a discussion of the difference between two different aspects of globalization: Global flows of people and global bridgeheads of people. Keyword Globalization, cultural clusters, film industry, Bollywood, Hollywood Acknowledgements Some of the underlying research was conducted in collaboration with Florian Täube during his employment at Tanaka Business School, Imperial College. The author is grateful for the kind collaboration of interviewees in Mumbai. They cannot all be listed here, but Rajesh Jog, Gurneeta Vasudeva, Satyadev Barman, and Raj Kaushal provided invaluable help. Thanks also to Kunal Singla, Erik Vinter, Zunia Ashan, Zanda Indriksone, and Christine Fur Poulsen for research assistance, and to Yuko Aoyama, Gernot Grabher, Mita Lad, and Allen Scott. The Danish Council for Strategic Research’s Creative Encounters project funded part of the research. Author Mark Lorenzen is Associate Professor for the Department of Innovation and Organizational Economics at Copenhagen Business School, Kilevej 14A, 3., DK- 2000 Frederiksberg, Denmark Mail: mark@cbs.dk, fax: +45 38152540 Page 2 / 38 Creative Encounters Working Paper # 26
Go West: The Growth of
Bollywood
“The barriers have been broken and the doors have been opened.”
Yash Chopra, Bollywood’s most prolific film producer
(interviewed in Mumbai, March 18, 2006)
Introduction
This paper builds insight into how globalization impacts cultural clusters
through a detailed case study of Bollywood, the cluster of film and media
companies in Mumbai, India.
Economic geography typically explains the growth of cultural clusters by virtue
of self-reinforcing external economies, and scholars point to early-mover
advantages of a few cultural clusters in Western economies, such as Los
Angeles, Paris, London, Milan, and so on. In this logic, globalization for the
cultural industries mainly entails westernization of mass culture and
subordination of cultural clusters in emerging economies to peripheral
positions in global networks. These clusters are seen as serving only local
markets or carrying out production outsourced from core clusters in the West.
Empirical evidence however begins to challenge such claims. Several cultural
clusters in emerging economies now grow fast, export their products to a wide
range of global markets, and occupy more and more central positions in global
networks. Unfortunately, empirical documentation of these clusters remains
sparse. Consequently, economic geography currently offers little insight into
how globalization impacts their growth.
In order to stimulate such insight, this paper reports the findings of a case study
of Bollywood. Producing roughly 1100 films annually, double that produced by
USA, India is the world’s largest film producer. Bollywood, with an estimated
3.6 billion tickets sold globally in 2001 (compared to Hollywood’s 2.6 billion), is
arguably one of the world’s most prolific cultural clusters (Kripalani and
Grover, 2002; Lorenzen, 2009). Bollywood is well suited for theory building
because it is an extreme and prototypical case of a cultural cluster that grows
under globalization. With its growing global impact upon films, music, dance,
and other art forms, Bollywood is developing its own strong global brand and
is also becoming big business, attracting massive investments. Whereas
Hollywood film producers and investors remain unable to make inroads into
India, Bollywood companies now export at a massive scale to USA and other
attractive consumer markets, and acquire cinemas and production companies
aboard. Using extensive triangulation of novel primary data, in the guise of one
quantitative and four qualitative data sets, the paper investigates why
globalization has a positive effect on Bollywood.
Page 3 / 38 Creative Encounters Working Paper # 26Cultural clusters, film clusters, and globalization Cultural clusters While not resting upon economic geography’s most impressive theoretical edifices, the problem of spatial clustering of specialized economic activity nevertheless attracts much scholarly attention. Markusen (1996); Feser and Bergman (2000); and Gordon and McCann (2000) seek to build taxonomies, and an increasing number of scholars also seek to theoretically explain the emergence and growth of clusters in cities and Marshallian districts. There is agreement that whilst every cluster typically emerges through an idiosyncratic confluence of historical factors (such as flagship firms, universities, capital, and politics), the continuous growth of a cluster hinges upon its ability to create cumulative causation (Myrdal; 1957) of one or several economic benefits for business firms of geographical co-location: External economies. External economies of scale are the benefits of co-location of firms undertaking very similar economic activities. They encompass attraction of specialized labor, learning among firms with cognitive proximity, plus small and flexibly specialized firms’ ability to offer high capacity and high quality through collaboration (Glaeser et al., 1982; Maskell, 2001, Morgan, 2004; Storper and Venables, 2004; Gordon and MacCann, 2005; Iammarinoa and McCann, 2006; Lorenzen and Frederiksen, 2008). External economies of scope encompass the benefits of co-location of diverse firms, be that firms in different stages of a value chain, or firms in entirely different industries. Such benefits encompass learning along value chains or across industries, facilitated by high trust and low transaction costs, and the opportunity for single firms to specialize deeply and build internal scale economies within single value chain activities (Henderson, 1988; Martin and Sunley, 2003). Survival of a cluster hinges upon its sustained competitiveness and growth. Hence, if it is not able to adapt its external economies to overcome exogenous (e.g. market) shocks or endogenous (organizational and political) challenges, the resulting lock-in to an obsolete development path will weaken and eventually erode the cluster (Martin and Sunley, 2006). With the “cultural turn” (for an overview, see Martin and Sunley, 2007), economic geographers dedicate increasing attention to a particular kind of cluster: Agglomerations of firms in cultural industries such as film, TV, music, advertising, publishing, and so on (Caves 2000; Throsby 2001; Hesmondhalgh 2002; Pratt and Hesmondhalgh 2008). Even if the majority of research has been content with describing prolific cultural clusters found in metropolises around the world (Scott and Power, 2004; Cooke and Lazzeretti, 2008), some scholars do offer explanations of their growth. Explanations of cultural cluster growth typically adhere to the arguments of external economies outlined above. For example, Grabher (2002), Pratt (2002), and Lorenzen and Frederiksen (2005) demonstrate how proximity lowers transaction costs and allow for external scope economies in development projects among clustered advertising, media, Page 4 / 38 Creative Encounters Working Paper # 26
and music films. Furthermore, Bathelt (2005) outlines how local information “buzz” can propagate external scale economies in the guise of learning among clustered media production companies. However, some scholars also point out that, due to the urban location of virtually all cultural clusters (Scott, 1997; Scott and Power, 2004: Scott, 2007; Lazzeretti et al, 2008), it is also necessary to include external economies related to urban location (Jacobs, 1961) in the explanation. These include external scope economies arising from co-location of a cultural cluster with urban venture capital or with clusters in other industries (Lorenzen and Frederiksen, 2008). They also include the unique ability of large cities to attract talent (Glaeser et al., 2001; Florida et al., 2008). Film clusters Like other cultural industries, the film industry also has a tendency to cluster in particular cities. The long list of film clusters encompass diverse cities such as Copenhagen, Seoul, Rome, Hong Kong, Paris, London, Kyoto, New York, Los Angeles, and Mumbai. Attempts at theorizing clustering in the film industry have centered on the case par excellence, Hollywood. Scott (2005) argues that Hollywood emerged in the first decades of the 20th century as a result of a coincidental confluence of royalty exemptions and external scope economies related to urbanization (land and finance) in Los Angeles. Then, the cluster developed external scale economies in the guise of huge specialized labor market and external scope economies by attracting a large number of different film value chain activities to Los Angeles. However, early on, taking advantage of its huge home market (Krugman, 1980; Davis and Weinstein, 2003), Hollywood also began to develop powerful internal economies. First, horizontal integration of production into large “studios” created huge scale advantages in production (Prag and Casavant, 1994; Eliashberg et al., 2006). Subsequently, widespread vertical integration of film finance, marketing, distribution and exhibition activities into these studios, forming a handful of powerful “majors”, created tight value chain control. The majors thus secured a relatively certain demand for their films (de Vany, 2004). After the Second World War, the cluster strengthened its external scale economies further. It built a flexibly specialized production model consisting of small-scale independent production firms, specialized suppliers, and freelancing creative and technical labor (Robins, 1993; Storper, 1989; and Anderson, 1987; Storper and Christopherson, 1987; Blair, 2001). The Hollywood case has dominated theorizing of film clusters to the extent that much research of other film clusters now takes for granted that these cannot compete against Hollywood, not even on their own home markets. This is because the development path of Hollywood rests on early-mover advantages (Scott, 2005). After its emergence, the film cluster moved so fast in building external and internal economies that it has since enjoyed dominance on global film markets (Vogel, 1998; Wasko, 2003; de Vany, 2004; Epstein, 2006). Bakker (2005) offers a theory about this: Hollywood introduced endogenous sunk costs (Sutton, 1991) into the film industry by investing heavily in production quality, Page 5 / 38 Creative Encounters Working Paper # 26
marketing intensity, and distribution (Aksoy and Robins, 1992; Wildman 1995; Wildman and Siwek, 1988). Thus, the cluster raised the film industry’s entry barriers far beyond what was given by industry-exogenous demand and technologies. Hollywood’s marketing and distribution system overpowered most competitors on any export market. As a result, the cluster has been able to overcome the liabilities of foreignness (Hymer, 1976) on export markets that cultural clusters generally face because consumers abroad have different preferences for languages and styles. Hollywood has created a general preference for English language films and Hollywood styles and narratives on most of the world’s markets for films (Hoskins and Mirus, 1988; Vogel, 1998; Elberse and Eliashberg, 2003; Papandrea, 1998; Doh, 2001; Lee and Waterman, 2006). Contrary to standard assumptions in industrial economics (Sutton, 1991), the film industry has been largely dominated by one cluster ⎯ even during the last half a century when globalization expanded the size of film markets hugely. Bakker (2005) argues that this is because the endogenous costs sunk by Hollywood have continued to limit the competitiveness of other film clusters. In the following, we shall take a closer look at globalization and what it may mean for cultural clusters in general and film clusters in particular. Globalization Internationalization can be understood as trade and other relations (such as agreements and alliances) between nations or nationally based firms and organizations. By contrast, globalization entails the integration of a multitude of nations, firms, and organizations into global economic, cultural, and to some extent also political, systems (Ohmae, 1990; Held et al., 1999; Daly 1999, Friedman 2000, Stiglietz 2002). To be more specific, during the last half a century, political shifts as well as developments in transportation and communication technologies have facilitated the emergence of global product markets, labor markets, capital markets as well as global institutions and organizations, such as the UN, numerous NGOs, and transnational corporations. A key dimension of globalization is connectivity: The most influential trait of global institutions, firms and organizations is not the fact that they are represented at multiple locations, but that they create global connections and networks (Amin, 2002; Sheppard, 2002; Dicken, 2003). Contrary to some claims (e.g. Cairncross, 1997; Friedman, 2000), globalization, and the transportation and communication technologies that facilitate it, does not mean the demise of spatial clustering (Morgan, 2004). It does entail global and hence tougher competition, but it also offers global market opportunities, and it empowers clusters by connecting them in global networks. Whether a cluster is able to survive under globalization is highly influenced by how it is connected in global networks. On the one hand, some clusters become subordinate suppliers to network hubs, more knowledge-based and value- adding clusters higher in the value chain (Humprey and Schmitz, 2002; Mudambi, 2008; Pratt, 2008). On the other hand, other clusters grow into hubs Page 6 / 38 Creative Encounters Working Paper # 26
themselves, by establishing pipelines of information from the world outside and boosting local innovation (Amin and Thrift 2002; McKinnon et al., 2002; Bathelt et al., 2004). Whether a cluster can profit from such pipelines depends upon its absorptive capacity (Gertler and Levitte, 2005; Giuliani, 2007). Many scholars still address emerging economies largely as passive recipients of products and investments from the OECD area. The bulk of the research on what globalization entails for cultural clusters located outside the dominant nations USA, UK, France, Italy, Japan, and so on, has focused on these clusters as peripheral in global networks. For example, Pratt (2008) considers the subordinate role of cultural clusters in emerging economies in global value chains, and Wasko and Erickson (2008) analyze how cultural production is outsourced in order to take advantage of cheap labor in such clusters. In cultural studies, a similar focus can be traced in the claims that a dominant Western cultural core homogenizes and westernizes its peripheries (Appadurai, 1996; Tomlinson, 1999; Giddens, 2000; Holton, 2000; Anheier and Isard, 2008). In the film industry, globalization has, for more than half a century, brought about a substantial growth of markets. Growing global trade and aligned consumer preferences for (Western) film narratives have allowed mass producers of films ⎯ mostly Hollywood, but UK and France have also played roles ⎯ to create and nurture a global mass market for mainstream films. During the last two decades, however, globalization has also facilitated the growth of global niche markets ⎯ such as art film aficionados, Kung Fu experts, or Manga lovers. Whereas global mass markets are stimulated by new technologies of film exhibition and distribution, global niche markets crucially hinge upon them. Largely, it is satellite TV, DVD, and the Internet that allow producers of niche films to reach their audiences around the world. Globalization has also meant that the world’s film clusters are subjected to tougher competition, and that some of them are becoming connected in global networks of co-production (for a discussion, see Morawetz et al., 2007). What does globalization, then, mean for the growth and survival of different film clusters? Summarizing the research, one finds a skepticism regarding the indigenous growth prospects of peripheral cultural clusters. Scholars point out that globalization allows Hollywood it to further consolidate its internal economies. Major film producers, now being huge multi-media conglomerates, invest further in marketing and distribution power, as well as use intellectual property rights from films across multiple platforms, such as TV, computer games, music, and online and print media (Scott, 2005; Epstein, 2006; Flew, 2007). Globalization also strengthens Hollywood’s external scale economies by allowing the cluster to attract talent from around the world. Finally, globalization allows Hollywood to outsource labor-intensive production phases to cheaper film clusters around the world. Consequently, most studies of film clusters outside the USA have focused on the impact of Hollywood “runaway” productions (Coe, 2001; Wasko, 2003; Coe and Johns, 2004; Vang and Chaminade, 2007), rather than understanding the indigenous growth of these film clusters. Page 7 / 38 Creative Encounters Working Paper # 26
Viewing the majority of the world’s cultural production as peripheral is of course unreasonable in an anthropological perspective. But it is also imprecise in an economic sense. Numerous cultural clusters in emerging economies now grow fast, export their products to a wide range of global markets, and occupy more and more central positions in global networks (Scott and Power, 2004; Cooke and Lazzeretti, 2008; Barrowclough and Kozul-Wright, 2006). In the film industry, notable examples encompass clusters in Mexico, Brazil, Korea, Taiwan, Nigeria, and, in particular, India. The empirical documentation of these film clusters remains sparse. As a result, little is still known about their growth dynamics, and how it is influenced by globalization. In order to build knowledge of the seemingly positive impact of globalization upon cultural clusters in emerging economies, this paper offers an analysis of the recent growth of the most prolific of them: Bollywood. In the following, I shall outline how the paper undertakes this analysis. Method The paper seeks to build new knowledge from empirical research, and the most appropriate strategy for that purpose is the case study method. In contrast to surveys and econometric methods, which allow for theory testing through correlating different phenomena, the case study method facilitates insights into the processes behind observed phenomena. Hence, it is suitable for building theory about their conditions and causes. The following sections explain how the case study of Bollywood was designed. Case Selection The purpose of the paper’s case study is not to test or illustrate incumbent theory, but to obtain new insights (Eisenhardt, 1989; Eisenhardt and Graebner, 2007). Hence, the paper undertakes an explorative case study, designed to allow for insights to emerge as a consequence of, not prior to, data collection (Yin, 1994). As mentioned above, the general problem the paper seeks insight to is how cultural clusters in emerging economies are influenced by globalization. However, it is not possible to undertake a multiple case study of cultural clusters in emerging economies: The total population of such clusters is too small to allow for probabilistic sampling (Stake, 1995), and resource constraints prevent undertaking a maximum variation sampling (George and Bennett, 2005). Instead, the paper undertakes a single case study. The research design that allows best for building insights on the basis of a single case is purposive selection of an extreme (outstanding) case, the findings for which can then be inferred logically to apply to a larger population of cases (Flyvbjerg, 2006). For example, positive findings for a negatively deviant (critical) case, can be inferred to also apply to less deviant cases. Or, when Page 8 / 38 Creative Encounters Working Paper # 26
particular contextual factors allow a case to stand out, findings for that
(paradigmatic or prototypical) case can be inferred to be revealing or illustrative
for mechanisms or tendencies that may apply to other cases if the contextual
factors also come to apply to them. All theory built from single cases may be
subject to later testing through multiple case studies or surveys (Eisenhardt and
Graebner, 2007).
Bollywood is picked as the paper’s single case, because it, rather than being
representative of all cultural clusters in emerging economies, is extreme and
prototypical. Bollywood is extreme in the sense that it is the cultural cluster ⎯
arguably, not just in the emerging economies, but on a global scale ⎯ that
performs best under globalization. Some insights into precisely how
globalization impacts Bollywood may, with caution, be inferred to less extreme
cultural clusters. The cluster is prototypical in the sense that India, with its
wide-open borders for flows of products, labor, and capital, is very intensely
subjected to globalization. Insights from Bollywood may hence be very
illustrative of the future for other cultural clusters in emerging economies.
Data sources and triangulation
In order to avoid problems of validity arising from data bias, we chose to
triangulate (Yin, 1994; Stake, 1995) one quantitative and four qualitative data
sets. These sets encompass aggregated data on industry and cluster level
(statistics, government reports, key informant interviews) as well as data on
changes on the single firm and inter-firm relational level (firm-level case studies
and interviews with managers and talent). Even if a longer historical backdrop
is given by the quantitative data sets, the main data sets focus upon the changes
in Bollywood during the period 2003-2007, where the cluster’s performance was
skyrocketing and changes in its external and internal economies became
apparent.
In order to build the qualitative data sets, a total of 58 interviews were carried
out. As Bollywood informants are unreachable through formal channels, the
author and collaborators had to use multiple personal networks as well as show
up unannounced at social and business events in order to arrange interviews.
Hence, four elongated periods of fieldwork in Mumbai during the period 2005-
2008 were needed. Interviews were all face-to-face, of durations between 30
minutes and 3 hours, and taped, transcribed, coded, and interpreted as
described below.
Figure 1 below gives an overview of the data sources.
Figure 1. Data sources
Data set Description of data Method of building data set
type
Page 9 / 38 Creative Encounters Working Paper # 26Quantitative data
Database on Names of production Primary data. We integrated data
all 11,500 company, distributor, on Hindi language films made
Bollywood cast, and other key available from Indian Motion
films participants of each Pictures Producers’ Association and
produced film. Indian Film Trade with data from
1913-2006 trade journals assembled by
ScreenWorld.
Qualitative data
Interviews Descriptions of Primary data. 21 interviews.
with 17 key changes at film We made a purposive sample of 6
informants production level and managers from the biggest or
cluster level (external otherwise most influential
economies and impact production companies plus 11
of globalization). informants from industry
associations, related industries,
and industry observers
(journalists, historians). As the
ongoing changes of Bollywood
became very conspicuous after
2003, the informants could
discuss these explicitly.
We used a short unstructured
protocol for open-ended
interviews (Miles and Huberman,
1984; Stake, 1995). The key
statements of each informant
were triangulated with
statements of other informants.
Government Descriptions and Secondary data. We made a full
reports on expert evaluations of sample of all reports published in
Bollywood development trends at the period 2005-2008:
industry and cluster CII and KPMG, 2005; FICCI and
levels. PWC; 2006; 2007; 2008; IBEF and
PWC, 2005; Khetepal, 2005; Kohi,
2003; Kohli-Khandekar, 2006;
USIBC and Ernst and Young,
2008.
30 embedded Descriptions of Primary data. 18 interviews.
case studies changes at firm level We purposively selected 30 films
and inter-firm as extreme cases: Those with the
relational level best box office performance ⎯
(production project not only representing the
design and commercial core of Bollywood
performance). during the period of our study,
Page 10 / 38 Creative Encounters Working Paper # 26but also likely to cause the
greatest level of imitation by
Bollywood films produced after
our study. Out of a total
population of the 652 Hindi
language films produced in 2003-
2005, a sample of the top ten
earners for each year (identified
by deducting production costs
(listed on
www.ibosnetwork.com) from box
office collections in the year of
release (listed on
www.ibosnetwork.com and
www.imdb.com) for the top 35
box office grossing films (all
territories) for each year) was
selected. This yielded a sample of
30 film projects, and interviews
with producers (who were often
also manager-owners) in 15
production companies were
undertaken, covering 23 of the
case films.
The interviews had a replicated
design with a semi-structured
500-word interview protocol.
Answers were checked for case-
specific influences (the
production company’s history
and external environment) and
triangulated using online
resources and the films
themselves.
Interviews Results and Primary data. 19 interviews.
with 15 interpretations from A sample of creatives and
random other data sources managers (2 actors, 7 directors, 2
informants were discussed. scriptwriters, 1 PR freelancer, 3
managers of production
companies) was randomly
selected through snowballing.
A semi-structured interview
protocol was used.
Page 11 / 38 Creative Encounters Working Paper # 26The data sources were triangulated as follows. In order to capture cluster emergence, growth and changes in external and internal economies, a novel original database on all Bollywood films spanning the period 1913-2006 was constructed. However, while such historical sources allow for representation of a historical change, insight into a change that is still ongoing and for which no statistics yet exist, hinges upon other data types (George and Bennett, 2005). In order to get such insight, we triangulated the statistics with interviews with 17 key informants with knowledge of Bollywood’s current growth and changes. In order to compensate for potential bias of these interviewees, we triangulated again, this time with expert evaluations of industry trends as published in government reports. In order to build detailed insights into the impacts on globalization upon single Bollywood companies and products, we used qualitative data in the guise of 30 case studies, embedded in the overall Bollywood case (Yin, 1994). The embedded cases were extreme: We picked the most successful film projects in the period 2003-2005, because these were likely to give the best insight into current dynamics of competition and the possible future ways of producing Bollywood films. These case studies added deep insights into e.g. the product development processes and changing external vs. internal economies in Bollywood. In order to compensate for the potential problem of the researcher’s bias when carrying out subjective interpretations of such cases (Eisenhardt and Graebner, 2007), triangulation of our interpretations was now made with two additional qualitative data sources. First, interpretations were triangulated with the aforementioned government reports. These supported, but added no new insight to our interpretations, and finally, we triangulated with (sometimes repeated) interviews with 15 randomly chosen creatives and managers. After 19 of these interviews, no challenges to the interpretations came forward, and consequently, further triangulation was halted (Lincoln and Guba, 1985; Yin, 1994). Data coding and interpretation As mentioned, the case studies were explorative, allowing for new insights to be built unconstrained by incumbent theory. Hence, a round of first-order open coding (Glaser and Strauss, 1967) of the data took place, without modifying data to fit existing theoretical categories. The results of this coding were subsequently coded again in order to add new knowledge to existing theory. In this coding, the new insights were, only as far as that made sense, interpreted using existing theoretical categories in economic geography. The categories used for this axial coding into second-order insights (Strauss and Corbin, 1990) were cluster emergence, external economies of scale and scope, internal economies, globalization, pipelines, and absorptive capacity. In the sections to follow, the paper presents the empirical results, organized using the above theoretical categories. The empirical material is presented in two parts. First, the largely descriptive section 4 provides a historical overview Page 12 / 38 Creative Encounters Working Paper # 26
of how Bollywood emerged in the early 20th century, and how its external and internal economies were very low up till the 1990s. At that time, a sudden export boom drove massive investments and put the cluster on a self- reinforcing positive development path. Section 5 then explains how Bollywood could leapfrog thus, by analyzing in detail how globalization impacts the cluster. Section 4 is based on secondary data (statistics and government reports) triangulated with key informant interviews, and section 5 presents results from interviews with managers and talent (the 30 embedded case studies plus interviews with random informants), triangulated with key informant interviews and secondary data (government reports). Background: The development of Bollywood This is the first of the paper’s two empirical sections. It rests on secondary data triangulated with key informant interviews, and describes the emergence and history of Bollywood, focusing on the cluster’s current remarkable transformation and growth. 20th century: Cluster emergence and low external and internal economies With its first production in 1913 and producing more than 100 films annually by 1933, the Indian film industry had an early start. The emerging film industry clustered in two cities. In the Indian cultural capital, Kolkata (then Calcutta), there were external scope economies of co-locating with cultural clusters of literature, music, and theatre. In the economic powerhouse of the Indian economy, Mumbai (then Bombay) film producers could enjoy external scope economies of venture capital from manufacturing industries and merchants spilling over into film production. The nascent Bollywood cluster produced both the first Indian film and the first Indian sound film in 1931. By then, more than half of India’s film production took place here. With the advent of sound films, film imports to India from Hollywood and other foreign film clusters virtually seized, and the home market segregated along language divisions, the biggest market segment being that of Hindi films (a language used across North India and today’s Pakistan). Even if Hindi was not the official language in Mumbai, centuries’ inflow of Hindi-speaking migrants to the city meant that it became India’s dominant Hindi film cluster. From its inception, Bollywood was a cottage industry (Shoesmith, 1987) with specialized production, distribution and exhibition companies enjoying only low external scope economies of collaboration, and had built no internal scale economies at all. Internal and external economies of the cluster grew in the 1930s and 1940s, albeit modestly. With rapid urbanization in North India, the market for Hindi films grew (and was soon to be boosted further when Hindi became India’s national language). However, because Indian cinema exhibition consisted of Page 13 / 38 Creative Encounters Working Paper # 26
thousands of small, single screen-cinemas with low ticket prices spread across the subcontinent, with very primitive technologies available to distribute films to them, it was not possible for Bollywood to build internal economies in either film exhibition nor distribution. The only value chain activity where internal scale economies were possible was production. Like in the contemporary Hollywood cluster, in the 1930s, Bollywood grew a handful of studios which by the early 1940s produced around two thirds of Bollywood’s 150-200 annual films (Barnouw and Krishnaswamy, 1980). However, these Bollywood studios did not have the money or incentive to integrate downstream. In stead, they relied on external scope economies arising from their location in Mumbai together with a growing network of hundreds of small-scale distributors, each of these competing to obtain distribution rights from the studios. Another economy of scope was the good relationship the studios built to financiers from other Mumbai industries, willing to channel capital to film projects. After Indian independence in 1947, Bollywood changed dramatically, and many of its internal and external scale economies eroded. The studios were weakened by poor sales and rising fixed costs during the Second World War. After the partition of India and Pakistan, dozens of small-scale film production companies that entered Mumbai from Pakistan outcompeted the studios altogether. Bollywood once again was completely horizontally and vertically disintegrated. The new production companies outcompeted the incumbents because they, like their contemporaries in Hollywood, developed a film formula, which, after proving its market value, would successfully be applied again and again with small variations. Together with the disintegrated structure, this formula would characterize Bollywood for the latter half of the 20th century. The Bollywood formula appealed to an all-Indian audience across regional, social, and religions and social divides. Hence, it was commercially extremely successful from the start and pursued and refined vigorously in the decades to follow. In the 1970s, it was endearingly named masala (Hindi for “spice mix), because it blended genre elements such as romance, drama and comedy with song-and-dance sequences in symbol-driven narratives. Productions were lavish and expensive, and a key component to masala was also the use of star actors, driving up production costs further. Having no internal scale, production companies could only engage in so artistically complex film projects because they were able to draw upon social networks among producers, directors, technicians, stars, and other creative talent (Lorenzen and Taübe, 2008). Having built such external scale economies in production, Bollywood nevertheless suffered from lack of external scope economies. Compared to the old studios, the new small production companies did not enjoy similarly good relations to financiers. Hence, for five decades, Bollywood saw constant struggle for film finance. Furthermore, relations between the many independent production and distribution companies became strained, raising risks for production companies. For distribution companies, low external scope economies meant a lock-in to a business model with inefficient releases and low collections. Page 14 / 38 Creative Encounters Working Paper # 26
Bollywood’s external as well as internal economies remained low for the rest of
the 20th century. A hit film would sometimes earn a substantial profit for a
production company, a distributor, or an investor. However, fortunes were
short-lived and exit rates high. Piracy also soared, due to inefficient
distribution, and is still estimated to cost Bollywood more than 30% of revenues
(Ernst and Young and USIBC, 2008). The high-profile Bollywood producer
Ashutosh Gowariker (interviewed in Mumbai June 24, 2005) described the
cluster in the second half of the 20th century very simply: “Chaos! Complete
mayhem!”
Late 20th century: Export growth
Since the 1930s, Bollywood had enjoyed a small but stable export to India’s
main trade partners (e.g. Russia and the Middle East), cultural neighbors (e.g.
Pakistan, Bangladesh, and Afghanistan), and countries where centuries of
migrations and British colonialism had created large Indian diasporas (e.g.
Nepal, Burma, South Africa, and Sri Lanka).1 Main exports of Indian films (all
languages) in 1988 were to The Arabian Gulf (35.4%); USSR (14%); Indonesia
(14.3%); Sri Lanka (3.8%); and Burma (3.6%). At that time, countries like
Morocco (3%); Jordan and Fiji (both 2.6%) were almost as big importers as
UK/Ireland (3.3%)(NFDC, 2007).
After Indian independence, new Indian diasporas began to grow rapidly in
countries offering education and work opportunities, such as USA, UK, Saudi
Arabia, United Arab Emirates, Canada, Singapore, Australia and New Zealand,
Kuwait and Oman. In 2001, the old and the new Indian diasporas combined
were conservatively estimated to a size of 20 million people globally. At that
time, in 11 countries, the Indian diaspora exceeded half a million people, and in
at least 48 countries, there were more than 10,000 Indian diaspora members
(Ministry of External Affairs, Government of India 2001; Walton-Roberts, 2004).
Given the rate of Indian immigration over the last decade, the diasporas in
some of these countries are now notably larger (see figure 2).
Figure 2. The size of the of the largest Indian diasporas (2001)
The new Indian diasporas The old Indian diasporas
USA: 1,700,000 Nepal: 4,000,000 ***
Saudi Arabia: 1,500,000 Burma: 2,900,000
UK: 1,200,000 * Malaysia: 1,665,000
United Arab Emirates: 950,000 South Africa: 1,000,000
Canada: 850,000 Sri Lanka: 860,000 ****
Singapore: 310,000 Mauritius: 715,000
Oman: 310,000 Trinidad and Tobago: 500,000
Kuwait: 295,000 Guyana: 395,000
Netherlands: 220,000 Fiji: 335,000
1
Even if the term “diaspora” earlier mainly referred to displaced Jews, it is now generally used about migrants from one particular
region or country, forming a permanent community in a new country of residence (Brah, 1996; Clifford, 1997; Fortier, 2000; Axel,
2002).
Page 15 / 38 Creative Encounters Working Paper # 26Australia: 190,000 Reunion: 220,000 Thailand: 150,000 Kenya: 105,000 Bahrain: 130,000 Yemen: 100,000 New Zealand: 100,000 ** Tanzania 90,000 The figures include non-resident Indian citizens (NRIs) and persons of Indian origin (PIOs), i.e. foreign citizens who held Indian passport earlier, or children of or spouses to Indian citizens or PIOs. All figures are rounded, as some are estimates from national statistical bureaus, other are based on census data. 2001 is the last year for which such a comparison of countries can be made. For instance, the last available survey from the UK (accessible from National Statistics UK, www.statistics.Gov.uk) is from 2001, and the last US Population Profile (available from US Census Bureau, www.census.gov) is from 2000. * In the UK, the 2001 sizes of the Pakistani and Bangladeshi diasporas were (rounded) 750,000 and 280,000, respectively. ** The New Zealand figure is from the 2006 census. *** The figure for Nepal varies widely, The 2001 census by the Nepali Central Bureau of Statistics (which can be accessed at www.cbs.gov.np) lists 100,000 NRIs in Nepal, but Non Resident Indians Online (nirol.com) estimates Nepali NRIs + PIO as high as 4,000,000, due to the historically fuzzy border between Nepal and India. **** The figure for Sri Lanka denotes only PIOs (”Indian Tamils”). Sources: National statistics as compiled by Ministry of External Affairs, Government of India (2001); 2001 UK Census (from National Statistics UK (www.statistics.Gov.uk); 2006 New Zealand census (from Statistics New Zealand, www.stats.govt.nz/census); 2001 Nepali census (from Nepali Central Bureau of Statistics, www.cbs.gov.np); Non Resident Indians Online (www.nriol.com, accessed April 10, 2008); 2001 Sri Lanka Census (from Department of Census and Statistics, Sri Lanka, www. www.statistics.gov.lk). From the 1990s, and at first largely inadvertently, Bollywood saw a sudden growth in export earnings from the new Indian diasporas in North America, the UK, and a range of Arab countries.2 Soon after, East Asian and markets such as Singapore and Australia began to rise. Exact export data cannot be provided, but two leading Bollywood distribution companies (of both film and TV) report that USA and UK now account for 50-60% of their export revenues (Ernst and Young, 2008). Today, Bollywood is the largest foreign exporter to the US entertainment market, and successful films are currently screened in up to 75 US cinemas, some earning in excess of USD 1 million in their opening weekend, making them appear in the top 20 box office charts (Times of India, 2006). The new Indian diasporas constitute very profitable export markets, because their purchase powers are much higher than those of the Indian home consumers or the old Indian diasporas. In 2005, UK’s GDP/capita was 9 times, 2 The Pakistani, Bangladeshi, Sri Lankan, and Afghani diasporas are also avid consumers of Bollywood products. In the UK, for instance, the Pakistani and Bangladeshi diasporas combined are almost as large as the Indian, in effect doubling Bollywood’s market in the UK. Page 16 / 38 Creative Encounters Working Paper # 26
Australia’s 14 times, USA’s 17 times, and UAE’s 20 times the Indian GDP/capita of 2,700 USD (CIA World Factbook, 2008). In USA, the highly skilled Indian diaspora is even more prosperous than the average population (Pandey et al., 2004; Mitra, 2007). This means that even modest sales on diasporic markets earn Bollywood large revenues. For example, subscribing to a Hindi content TV channel costs 24 USD monthly in the UK, six times the price in India (FICCI and Ernst and Young, 2008), and the cost of a video CD from leading distributors is typically 2.50 USD in India compared to 13-18 USD in USA for the same film on a DVD (www.shemaroo.com; www.erosentertainment.com; www.yashrajfilms.com). The revenues from theatre distribution abroad is even higher: The 2005 average cinema ticket price was 35 cents in India, 6.55 USD in USA, 7.8 USD in Australia, 8.17 USD in the UAE, and 9.55 USD in the UK)(European Audiovisual Observatory, 2007). As a result of its booming exports to these new profitable markets, Bollywood exports earnings have grown 30-50% annually during the last five years to its present 16% of total revenues (CII and KPMG, 2005). A recent estimate is that Bollywood now collects more than USD 200 million annually abroad, and a further export growth of over 20% is expected until 2010 (CII and KPMG, 2005; FICCI and Ernst and Young, 2008). Zee TV, India’s biggest Hindi TV channel, has 4 million subscribers abroad, collecting 40% of its revenue from exports (FICCI and Ernst and Young, 2008), and the leading cable provider in USA, Comcas, is now offering a Bollywood on Demand service. Bollywood’s collections from cinema exhibition on export markets has experienced a 21% growth in the period 2006-2007, now accounting for 9% of Bollywoods total turnover (FICCI and Ernst and Young, 2008). Present time: Investments in internal and external economies Earnings from the export boom since the 1990s have begun to change Bollywood profoundly. The first change is that incumbent companies are investing in internal economies. A couple of production companies that first stumbled upon big export successes channeled their earnings into building internal scale in production as well as developing internal distribution capabilities. The most successful of these, the grand old family-owned Yash Raj Films, invested its export earnings wisely and managed to reinforce its export success throughout the 1990s and the beginning of the new millennium. Now, this company has the largest turnover Bollywood has yet seen (100 million USD in 2005) and is fully self-supplying with production as well as most distribution services. A dozen of other incumbent production companies with just a few export successes under their belt are now also investing in internal economies in order to imitate Yash Raj Film’s business model. Furthermore, the biggest two of Bollywood’s many independent distribution companies, Eros and Shemaroo, are also investing rapidly in building a scale and efficiency hitherto unseen in Bollywood. The second change resulting from the export boom is a major policy shift. After only recently having opened the Indian borders to globalization in 1991, Indian Page 17 / 38 Creative Encounters Working Paper # 26
policymakers became convinced by Bollywood’s somewhat surprising export growth that the cluster could potentially become an important Indian growth factor, if given better framework conditions. This political shift was remarkable, because in stark contrast to many Western cultural clusters, Bollywood never enjoyed public support earlier. On the contrary, for almost a century, British colonial rule and later India’s manufacturing-based growth strategy impeded the film industry’s performance. But towards the end of the 1990s, Indian state governments began cutting the whopping entertainment taxes (in some states amounting to over 150% of ticket prices). Furthermore, the Indian national government, in a bid to encourage investors to engage with Bollywood, credited Indian entertainment industries “official” status with the semi-public Industrial Development Bank of India. This new policy framework worked to reinvigorate external economies of scope between Bollywood and financiers from other Mumbai industries. But it happened on a much larger scale than in the 1930s and 1940s: Tremendous investments began to flow from Mumbai’s booming telecom, software and media industries into strengthening Bollywood’s external and internal economies. The third major change of Bollywood is, therefore, new entries into the cluster from other industries. The most notable examples are the entry of four prolific Mumbai companies from other industries into Bollywood: Sahara One, the main Hindi TV channel; UTV, India’s biggest TV program producer, Percept, India’s biggest PR company, and Reliance, India’s biggest telecom company have ventured into film production and distribution on a big scale. All these newcomers to Bollywood focus on building internal economies in distribution, and rely mostly on external economies for content, buying most of the films they release from small independent production companies. Together with the two incumbent distribution firms that now build internal economies, these newcomers represent a new scale-intensive way of distributing films that is quickly winning market shares from the small-scale independent distributors. For the first time in Bollywood history, distribution of firms systematically create crowding effects, pre-empt piracy, and aggressively use new exhibition platforms such as video, satellite and pay TV, and the Internet ⎯ in fact, to a much higher extent than Hollywood (Currah, 2007). The last major change of Bollywood is not driven directly by the cluster’s export success in the 1990s, but it is facilitated by the policy shift that was a direct consequence of the export boom. Large investments now flow into Indian film exhibition. After decades of deteriorating and closing single-screen cinemas, new multiplex cinemas are being constructed in a large number of Indian cities. These investments aim at profiting from demographic developments on the Indian home market: The emergence of a 300+ million people strong middle class (growing at 5% annually), a development driven by the doubling over the last two decades of the Indian GDP/capita and the explosion of Indian cities. The growing urban middle-class audience prefers multiplex cinemas with greater choice and better amenities. More importantly, it has high purchase power and a high consumption of entertainment. Page 18 / 38 Creative Encounters Working Paper # 26
The growth efffects of thee ongoing investmen
i ts in Bollyw
wood’s extternal and
interrnal econommies are suubstantial. Entertainm ment is now, after IC
CT, India’s
fasteest growingg sector. Thhe revenuees of the In
ndian film industry h have grown n
360%% in the peeriod 1998--2005, and 58% in thee period 20001-2005. Bollywood
speaarheads thiis positive developm ment. In 200 06, Bollywo ood was accknowledg ged
by innvestors ass one of Ind
dia’s central growth industries. In 2004, B Bollywood
prodduced 16% of all Indiian films, but
b accoun nted for mo ore than 400% of the
Indiaan film ind
dustry’s revenues (CIII and KPM MG, 2005). 2006-20077 revenue
growwth for Bolllywood wasw 14% (ass compared d to the Ho ollywood ggrowth of 7%
7
for the same timme period), but the annual
a groowth rate could reach h as high ass
30% by 2010 (K Kapoor, 2005; FICCI and a PWC, 2007). Forr 2007, a reccord outpu ut of
257 films
f saw more
m ambiitious distrribution annd marketiing schemees than eveer
befo
ore. As illusstrated in figure
f 3 beelow, the most
m notablle developmment of
Bollyywood sin nce the late 1990s is noot producttion of morre films, bu
ut rather a
growwing horizontal integ gration of distribution
d n (the growwing perceentage of fiilms
releaased by thee top 5 com
mpanies) acccompanieed by dram matic turnoover growth h.
Figure 3. Bollyw
wood’s turn
nover, produ
uction, and horizontal
h i
integration 1998-2007
7
Turnnover and film produ
uction on the
t left axiss, top 5 com
mpanies’ reeleases on the
rightt axis.
Sourrces: Own database; FICCI
F and
d PWC (200 08)
As mentioned,
m , since the 1930s, Holllywood haas had veryy little pen
netration in
n
Indiaa. Today, Hollywood
H d has a maarket share of 4% in In
ndia, the loowest
anywwhere in thhe World (CII and KP PMG, 20055). Since the turn of thhe millennnium,
inveestors from
m Hollywoo od have sought to profit from thhe growin ng Indian
dommestic mark ket in otherr ways: Th
hrough ven nturing intoo co-produ uctions of
Hind di films an
nd purchasing shares in Bollywood compaanies. Not knowing how h
to ch
hurn out a good masaala film, ennjoying no social
s netw
work relations to starrs
Page 19
1 / 38 Creative Encounters Working Paperr # 26and other creative talent (Lorenzen and Täube, 2008), and with Bollywood not needing foreign venture capital, Hollywood’s attempts of entering India have been unsuccessful. By contrast, Bollywood is now strong enough (even with the ongoing financial crisis) to invest in USA. In 2008, Bollywood company Reliance placed a 500 million USD investment in Hollywood flagship Dreamworks. As we have seen in this section, towards the end of the 20th century, an export boom drove massive investments in Bollywood, and an evolutionary path of self-reinforcing external and internal economies seems now to be under creation. Why did this happen? How could Bollywood, which at the beginning of the 1990s was characterized by horizontal disintegration and no export infrastructures, boost its exports so dramatically? The next section will demonstrate that the answer is, simply put, globalization. Analysis: Globalization’s impact upon Bollywood Resting on interviews with managers and talent, triangulated with key informant interviews and secondary data, this section analyses in detail how globalization impacts Bollywood. The chapter revolves around the most important aspect for Bollywood of globalization: The new Indian diasporas. Accounting for the interaction between Bollywood and these diasporas, the paper analyses how it gives rise to growing exports, development of new products, and the cluster’s investments in marketing, distribution and exhibition infrastructures. How the diasporas strengthen Bollywood’s external and internal economies through exports As mentioned in section 2, cultural clusters generally face liabilities of foreignness on export markets (Hymer, 1976). Until recently, Bollywood suffered greatly from such liabilities. The cluster could not afford to invest in marketing and distribution on a scale sufficient to influence consumer on export markets to favor the distinctive masala film formula. However, Bollywood’s liabilities are much lower on particular segments of the most attractive global markets in USA, UK, Canada, and the Middle East: The Indian diasporas. Like other diasporas, rather than realizing “the myth of return to the homeland” (Brah, 1996; Clifford, 1997; Fortier, 2000), the Indian diasporas act out their longing for the homeland through symbolic acts, such as consumption of cultural products. And to a high extent, the Indian diasporas share the Indian home market’s appetite for Bollywood products. However, during most of the 20th century, Bollywood did not take advantage of the diasporas, because the cluster was unable to distribute its products to them on any significant scale. For example, in spite of its significant size, the diaspora in North America is very geographically dispersed and consequently, except Page 20 / 38 Creative Encounters Working Paper # 26
from in a few metropolises, cinema exhibition was generally unprofitable.3 Hence, until the 1990s, Bollywood treated all export markets as one, largely unattractive, territory. Most Bollywood production companies let the state- owned National Film Development Corporation handle all export rights, which were then acquired by independent small agents living abroad and focusing on distributing to only one national market with very little (if any) marketing effort. Consequently, in the West, Bollywood was consumed in a system parallel to that of mainstream Western films: Video rentals from ethnic shops and occasional screenings in peripheral and often run-down theatres or community halls (Brosius, 2005; Dudrah, 2006). For Bollywood film producers, foreign collections were meager, even for the relatively successful films. The prolific producer Yash Chopra (interviewed in Mumbai, March 18, 2006) recalls: “In those days, we used to release films for free. I had gone for London to attend the premiere and it was free. Nobody was supposed to pay us money. We used to pay our own costs to promote the movie. They would maybe send me 2-3 tickets, otherwise no money to us.” Globalization has completely changed that situation. During the 1990s, Bollywood was less daunted by property rights complications than Hollywood (Currah, 2007) and began to embrace new technologies of exhibition and distribution in order to reach markets abroad. Bollywood’s use of TV (cable and now also satellite), video (earlier VHS, now DVD), and, lately, Internet downloading, rendered the problem of the diasporas’ geographical distribution much less important. As soon as Indian satellite TV channels became accessible, the diasporas responded enthusiastically and fast, and when Bollywood products were offered on video and the Internet, exports to the diasporas grew steadily. Diasporic demand has risen so fast that today, all main Bollywood distributors, as well as a range of production companies, sell both to TV channels abroad and directly to consumers through DVD retailers and a plethora of websites. Says top producer Yash Chopra (interviewed in Mumbai, March 18, 2006) about the use of new distribution technologies: “It’s the start of a revolution if you ask me … Indian films are going global.” Veteran producer Boney Kapoor (interviewed in Mumbai, March 30, 2006) adds: “The breakdown [for my films] today would be 40% Indian box office, the rest overseas and electronic [distribution]”. As described in section 4, the export boom allowed incumbent Bollywood film producers to build their own distribution capabilities, and created new scale- intensive distribution companies. The cluster’s new distribution muscle is also used for aggressive marketing: For the first time in a century, Bollywood products are now marketed across the world, using all possible platforms. Even if this marketing infrastructure is being developed in order to target the diasporas, it also earns Bollywood a presence in mainstream video stores and on film festivals. As a result, mainstream European and US TV channels, film festivals, and cinemas are now beginning to purchase Bollywood films. 3In UK, the Indian/Pakistani diasporas agglomerate more notably in cities with manufacturing and low-skill jobs. Page 21 / 38 Creative Encounters Working Paper # 26
You can also read
